Say, I have a numpy array like this:
[[ 1 2 3 5]
[ 4 5 6 8]
[ 7 8 9 11]]
I want to get the sum of (col 0+col 2) and (col1+col3) for each row. I know its probably elementary, but cant get it. Kindly help
Add a comment
|
3 Answers
Well yes it's pretty elementary once you know the assignment scheme of numpy arrays :
Let
x = [[ 1 2 3 5]
[ 4 5 6 8]
[ 7 8 9 11]]
y1 = x[:,0]+x[:,2] # Sum of columns 0 and 2
y1 = array([4,10,16])
y2 = x[:,1]+x[:,3] # Sum of columns 1 and 3
y2 = array([7,13,19])
1 Comment
Ash
So what if I have 1000 columns? Should we write 2000 lines of code for that?
This should do the trick:
import numpy as np
mat=np.arange(12).reshape(3,4)
A=np.sum(mat,axis=0)[0::2]
B=np.sum(mat,axis=0)[1::2]
A = (col 0+col 2)
B = (col1 + col3)
1 Comment
chron0x
Note, that I used the variable mat, which is different from your matrix. Just substitute mat in line 3 and 4 with the variable of your matrix.
Setup
Out[99]:
array([[ 0, 1, 2, 3],
[ 4, 5, 6, 7],
[ 8, 9, 10, 11]])
Solution
#reshape the array to 2 columns, then sum columns, finally reshape it back to 2 columns.
a.reshape(-1,2).sum(1).reshape(-1,2)
Out[102]:
array([[ 1, 5],
[ 9, 13],
[17, 21]])
